Preliminary data inspection

from above output we can infer that mostly values are which are null is in one coloumn i.e. Cuisines while as Resturant Name has 1 value which is missing which can be ignored. Hence we have to deal with null values of cuisines and make sure it does not affect our sample data

It can be infer from above that there are no duplicates in the dataset

Performing EDA Week 2

Hence Top 10 Cusines are North indian, Chinese, Fast food, Mughlai, Italian, Bakery, Continental, Cafe, South Indian and Desserts

Hence , the most served cuisine across the restaurant for each city can be seen in above table. From that we can infer that North Indian is a popular cuisine in New Delhi.

Approx 85% of the restuarants have a distribution cost for two is 0- 25 in terms of dollars. Other resturants are expensive in terms of average cost for two

Around 26% of the resturants have around 3.5 rating and 1200 restaurants which have 4 rating above are popular among people which can be infer from above Histogram

From above heatmap, we can infer that price range, vote, country code are the factors which have effect on ratings of a particular restuarant

Use One Hot Encoding to Get More Information about the Venue Categories

These above factors like Has table booking, delvery options and available cuisines have direct and indirect effect on recommending a star restuarnts. From an interective dashboard an user can interect according to the locality in order to find out the aggregate rating of the restaurant and accordingly choose a star resturant in order to visit the same restaurant.

Factors like cost, delivery and table booking affects one restaurant rating, hence each city has one star resturant in a particluar cuisine. With preference and demand of particular cuisine or services offered by restuarnts the choice of star restaurant in a particular city changes.